Testing Methodology.

At DexMouse, we believe subjective feelings about a mouse are important, but they must be grounded in objective hardware data. Here is exactly how we extract that data.

Click Latency (NVIDIA LDAT)

We use a modified NVIDIA Latency Display Analysis Tool (LDAT). We wire the LDAT directly to the mouse switch contacts (bypassing the acoustic microphone method, which introduces variance). We measure the time delta between the electrical circuit closing and the signal registering in the OS. We take 100 samples per mouse and report the average.

Sensor Analysis (MouseTester v1.5)

We utilize a motorized linear rail to ensure consistent physical speed during testing.

• CPI Deviation: We measure actual counts generated vs expected counts over a fixed distance (10cm) at 400, 800, 1600, and 3200 CPI. Deviation > 5% is noted as a negative.
• Polling Stability (xCount vs Time): We plot sensor reports. An ideal plot shows points clustered tightly along a curve. Outliers indicate dropped packets or inconsistent MCU timing, which translates to micro-stutters in-game.

Build Quality & Tolerances

We evaluate shell rigidity by applying deliberate lateral pressure to the side walls. We check for pre-travel (distance the button moves before actuating the switch) and post-travel (distance the button moves after actuation) on all main clicks.